Abstract:We present a computational approach that incorporates a convolutional neural network (CNN) for detecting internal delamination in a layered 2D plane-strain anisotropic composite structure of transient elastodynamic fields. The two-dimensional spectral element method (SEM) is utilized to simulate the propagation of elastic waves in an orthotropic solid sandwiched by isotropic solids and their interaction with the internal delamination cavity. This work generates training data consisting of input-layer features (i.e., measured wave signals) and output-layer features (i.e., element types, such as void or regular, of all elements in a domain). To accelerate training data generation, we utilize explicit time integration (e.g., the Runge–Kutta scheme) coupled with an SEM wave solver. Applying the level-set method additionally avoids having to perform an expensive re-meshing process for every possible configuration of the delamination cavities during the data-generation phase. The CNN is trained to classify each element as a non-void or void element from the measured wave signals. twoClusters of identified void elements reconstruct targeted cavities. Once our neural network is trained using synthetic data, we analyze how effectively the CNN performs on synthetic measurement data. To this end, we use blind test data from a third-party simulator that explicitly models the traction-free boundary of cavities for anisotropic materials without the application of the level-set method. Our numerical examples show that our approach can effectively detect the internal cavities in an anisotropic structure made of aluminum and carbon fiber-reinforced epoxy using the measured elastic waves without any prior information about the twocavities' locations, shapes, and numbers. The presented method can be extended into a more realistic 3D setting and utilized for the nondestructive test of various anisotropic composite structures.

Inverse characterization of composites using guided waves and convolutional neural networks with dual-branch feature fusion

Application of Inversion Algorithms with Neural Network in Ultrasonic Measurement of Unidirectional Fiber Composite Plates

Inverse of Initial Stress in Carbon Fiber Reinforced Polymer Laminates Using Lamb Waves and Deep Neural Network

Determination of the Elastic Constants of A Composite Plate Using Wavelet Transforms and Neural Networks

Quantitative reconstruction of defects in multi-layered bonded composites using fully convolutional network-based ultrasonic inversion

Solving Inverse Wave Scattering with Deep Learning

Deep learning for full-field ultrasonic characterization

A variational approach to the inverse imaging of composite elastic materials

Inverse Design of Photonic Crystal Waveguides Using Neural Networks and Dispersion Optimization

Convolutional neural network and level-set spectral element method for ultrasonic imaging of delamination cavities in an anisotropic composite structure

Damage recovery in composite laminates through deep learning from acoustic scattering of guided waves

One-Dimensional Elastic and Viscoelastic Full-Waveform Inversion in Heterogeneous Media Using Physics-Informed Neural Networks

Resolving engineering challenges: Deep learning in frequency domain for 3D inverse identification of heterogeneous composite properties

Physics-informed and graph neural networks for enhanced inverse analysis

Full waveform inversion guided wave tomography with a recurrent neural network

Imaging of Fiber Waviness in Thick Composites with Unknown Material Properties Using Probability-Based Ultrasound Non-Reciprocity

Deep Convolutional Neural Network for the Inverse Design of Layered Photonic Structures

Ultrasonic signal classification and imaging system for composite materials via deep convolutional neural networks

Multichannel Crossed Convolutional Neural Network for Combined Estimation of Cortical Thickness and Bulk Velocities Using Ultrasonic Guided Waves: A Simulation Study

Inverse Design Framework With Invertible Neural Networks for Passive Vibration Suppression in Phononic Structures

Deep neural networks for the prediction of the optical properties and the free-form inverse design of metamaterials